Blog of Peter Lambert

In the news I saw a reference to this paper on recyclable polymers¹. The basic idea described is that these researchers have developed a class of polymers which allow efficient recycling. By combining the monomers of a bis-triketone and a di-amine a polymer is formed. This polymer can be broken back down to the monomers simply by soaking in acid.

The important advancement here is that once the starting monomers are regenerated they can be purified from any additives, and cleaned from impurities from the use of the polymer, and then used as starting materials for the generation of new polymer. This allows the recycled product to be just as high quality as the original product.

Let's take a look at the chemical reactions going on here. First, when they say they are using a triketone it is specifically a molecule with a carbon surrounded by three carbonyl moieties, an example is shown in Figure 1. A triketone is more appropriate than a normal ketone for this sort of reaction because it has a fairly acidic proton on that central carbon, the acidity is caused by stabilization of the resulting anion by the surrounding carbonyl groups, which can be represented using resonance structures as shown in Figure 2.

The reaction creating the polymer is the reaction of a triketone with an amine, which forms a diketoimine and releases a molecule of water, as shown in Figure 3.

This reaction can be reversed by hydrolysis under acidic conditions to regenerate the original triketone and amine, as shown in Figure 4.

The formation of a polymer would use a diamine and a bis-triketone as starting materials. When mixed, they form a polymer, as shown in Figure 5. Here I have shown the reaction using 1,2-diethylamine and the bis-triketone derived from adipic acid.

The reaction shown above would result in in linear polymer. A branching or networked polymer can be created using a triamine, such as the one shown in Figure 6.

The physical properties of the resulting polymer can be controlled by the choice of monomers. A mixture of diamine and triamine monomers could be used to tailor the polymer for a specific use. Instead of using the aliphatic bis-triketone derived from adipic acid, one could start with the aromatic bis-triketone derived from terephthalic acid, which is shown in Figure 7. This would give more rigidity to the resulting polymer. Alternately, one could form a polymer with less rigidity by using a bis-triketone derived from a longer chained di-acid, such as dodecanedioic acid.

The real advantage this polymer presents is that additives such as colorants, platicizers, or inorganic fibers for strength can be added without reducing the recylability of the polymer. Once the polymer is broken down by acid, the monomers can be purified using regular solution purification techniques.

I also wanted to call out at the formation of the bis-triketone, because I think the chemistry here, while not new in itself, is interesting as an example of organic synthesis. They start with two equivalents of dimedone² and one equvalent of adipic acid³. These were dissolved in methylene chloride with three equivalents of DMAP⁴, which will deprotonate both the adipic acid and the dimedone. They then added a solution of DCC⁵ in methylene chloride, which accepts the oxygen from the adipic acid, forming N,N'-dicyclohexylurea and allowing the coupling of the adipic acid and dimedone. The N,N'-dicyclohexylurea precipitates and was removed by filtration, and then the DMAP was removed by washing with hydrochloric acid. This scheme is shown below in Figure 8⁶.

1. Peter R. Christensen, Angelique M. Scheuermann, Kathryn E. Loeffler, Brett A. Helms, Closed-loop recycling of plastics enabled by dynamic covalent diketoenamine bonds, Nature Chemistry, volume 11, pages 442–448 (2019). [↩]
2. 5,5-Dimethyl-1,3-cyclohexanedione [↩]
3. Hexanedioic acid. You need 2 eq. of dimedone because the reaction adds one dimedone to each end of the adipic acid. [↩]
4. Dimethylaminopyridine, a base. You might think this would require four equivalents, but the DCC acts as a base as well so you don't need so much [↩]
5. Dicyclohexylcarbodiimine [↩]
6. I admit that I am being lazy about arrow pushing electrons here, but I think this shows the important part of the reaction. [↩]